Feeds:
Posts

## Water vapour feedback is simply written into climate models as parameters?

Over in another article, a commenter claims:

..Catastrophic predictions depend on accelerated forcings due to water vapour feedback. This water vapour feedback is simply written into climate models as parameters. It is not derived from any kind simulation of first principles in the General Circulation Model runs (GCMs)..

I’ve seen this article of faith a lot. If you frequent fantasy climate blogs where people learn first principles and modeling basics from comments by other equally well-educated commenters this is the kind of contribution you will be able to make after years of study.

None of us knowed nothing, so we all sat around and teached each other.

Actually how the atmospheric section of climate models work is pretty simple in principle. The atmosphere is divided up into a set of blocks (a grid) with each block having dimensions something like 200 km x 200km x 500m high. The values vary a lot and depend on the resolution of the model, this is just to give you an idea.

Then each block has an E-W wind; a N-S wind; a vertical velocity; temperature; pressure; the concentrations of CO2, water vapor, methane; cloud fractions, and so on.

Then the model “steps forward in time” and uses equations to calculate the new values of each item.

The earth is spinning and conservation of momentum, heat, mass is applied to each block. The principles of radiation through each block in each direction apply via paramaterizations (note 1).

Specifically on water vapor – the change in mass of water vapor in each block is calculated from the amount of water evaporated, the amount of water vapor condensed, and the amount of rainfall taking water out of the block. And from the movement of air via E-W, N-S and up/down winds. The final amount of water vapor in each time step affects the radiation emitted upwards and downwards.

It’s more involved and you can read whole books on the subject.

I doubt that anyone who has troubled themselves to read even one paper on climate modeling basics could reach the conclusion so firmly believed in fantasy climate blogs and repeated above. If you never need to provide evidence for your claims..

For this blog we do like to see proof of claims, so please take a read of Description of the NCAR Community Atmosphere Model (CAM 4.0) and just show where this water vapor feedback is written in. Or pick another climate model used by a climate modeling group.

This is the kind of exciting stuff you find in the 200+ pages of an atmospheric model description:

From CAM4 Technical Note

You can also find details of the shortwave and longwave radiation parameterization schemes and how they apply to water vapor.

Here is a quote from The Global Circulation of the Atmosphere (ref below):

Essentially all GCMs yield water vapor feedback consistent with that which would result from holding relative humidity approximately fixed as climate changes. This is an emergent property of the simulated climate system; fixed relative humidity is not in any way built into the model physics, and the models offer ample means by which relative humidity could change.

From Water Vapor Feedback and Global Warming, a paper well-worth anyone reading for who wants to understand this key question in climate:

Water vapor is the dominant greenhouse gas, the most important gaseous source of infrared opacity in the atmosphere. As the concentrations of other greenhouse gases, particularly carbon dioxide, increase because of human activity, it is centrally important to predict how the water vapor distribution will be affected. To the extent that water vapor concentrations increase in a warmer world, the climatic effects of the other greenhouse gases will be amplified. Models of the Earth’s climate indicate that this is an important positive feedback that increases the sensitivity of surface temperatures to carbon dioxide by nearly a factor of two when considered in isolation from other feedbacks, and possibly by as much as a factor of three or more when interactions with other feedbacks are considered. Critics of this consensus have attempted to provide reasons why modeling results are overestimating the strength of this feedback..

Remember, just a few years of study at fantasy climate blogs can save an hour or more of reading papers on atmospheric physics.

### References

On the Relative Humidity of the Atmosphere, Chapter 6 of The Global Circulation of the Atmosphere, edited by Tapio Schneider & Adam Sobel, Princeton University Press (2007)

Water Vapor Feedback and Global Warming, Held & Soden, Annu. Rev. Energy Environ (2000) – free paper

Radiative forcing by well-mixed greenhouse gases: Estimates from climate models in the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (AR4), WD Collins et al, JGR (2006)

### Notes

Note 1: The very accurate calculation of radiation transfer is done via line by line calculations but they are computationally very expensive and so a simpler approximation is used in GCMs. Of course there are many studies comparing parameterizations vs line by line calculations. One example is Radiative forcing by well-mixed greenhouse gases: Estimates from climate models in the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (AR4), WD Collins et al, JGR (2006).

### 126 Responses

1. on November 5, 2017 at 10:30 pm | Reply Timothy Spiegel

As for climate fantasy blogs…..the whole notion of climate modeling basics is pure fantasy. The models act as if radiation is the main driving force in heat transfer. It’s not. CO 2 and water vapor collide with nitrogen and oxygen conducting their captured radiation into said molecules before they ever have a chance to emit radiation back to the ground which is the foundation of the global warming theory.

• Timothy,

CO 2 and water vapor collide with nitrogen and oxygen conducting their captured radiation into said molecules..

This is correct and is believed by everyone working in atmospheric physics. You find it in textbooks, papers and also it is the basis of the way radiation works in climate models.

Thermalization of absorbed radiation by collision.

..before they ever have a chance to emit radiation back to the ground which is the foundation of the global warming theory.

Wow. Where did you learn that this erroneous idea is the “foundation of the global warming theory?”

From a paper? From a textbook?

Or from a fantasy climate blog?

• You can see this explained in the derivation of the equation of radiative transfer – Understanding Atmospheric Radiation and the “Greenhouse” Effect – Part Six – The Equations.

If you read through you see that the emission of radiation at a given wavelength is only dependent on the temperature of the local gas and the emissivity at that wavelength.

Usually we get visitors who haven’t read textbooks showing up and claiming the opposite – that when a CO2 or water vapor molecule absorbs a photon it re-emits it. And that the energy absorbed is not absorbed into the average temperature of the local gas by collision. Then we have to explain the average time for collision in gas at a given pressure is 1000x shorter than the time to re-emission of a photon for this gas molecule in an excited state, etc, etc.

I’m fine with people being confused. Atmospheric physics requires study. I’m constantly amazed at the huge confidence of people claiming “climate science says..” completely the opposite of what is actually in books and papers.

And I can’t remember anyone saying “oh, wow, that’s a surprise, I was wrong“.

It’s probably happened once or twice, but drowned out in my memory by the thousands of comments by hundreds of over-confident visitors.

• Timothy,

And if you take a look at The Amazing Case of “Back Radiation” – Part Two you will this graph where theory matches experiment:

This is the calculated and observed spectrum at the surface looking up. There are many papers with these kind of comparisons. You can see one from the top of atmosphere, looking down, over at Two Basic Foundations:

Sharp-eyed observers will wonder how the theory matches.

• Timothy Spiegel,

So how is it possible then to acquire emission spectra of the atmosphere or obtain a temperature reading of the atmosphere using an inexpensive IR thermometer?

The reason is that exactly as many ground state CO2 and H2O molecules are raised to an excited state by collisions with nitrogen and oxygen as excited state molecules are de-excited by collisions. Only a small fraction of collisions, however, result in emission of a photon. That’s pretty much the definition of local thermal equilibrium, which applies to most of the atmosphere. But there are lots of collisions, so there are still lots of photons emitted. As a result, the number of molecules in a given volume in the excited state is a function only of the local temperature and the energy of the excited state. Emission is proportional to the number of molecules in the excited state. The proportionality constant is the inverse of the lifetime of the excited state.

• Timothy wrote: “CO2 and water vapor collide with nitrogen and oxygen conducting their captured radiation into said molecules before they ever have a chance to emit radiation back to the ground which is the foundation of the global warming theory.

Many alarmists have over-simplified the theory of global warming to the concept that GHGs “trap” heat in the atmosphere by absorbing outgoing thermal infrared. You are correct to dismiss the idea that heat is “trapped” by GHGs – absorbed photons are “thermalized” by collisions as you describe.

Nevertheless, the surface of the planet emits an average of 390 W/m2 of thermal IR and only 240 W/m2 reaches space. Clearly the absorption of thermal IR by GHGs slows down the RATE at which heat reaches space despite “thermalization”. “Slowing down” is different from trapping.

One simple explanation is that the presence of more GHGs in the atmosphere means that the average photon escaping to space must be emitted from higher in the atmosphere to escape past more GHGs. In the troposphere, the higher the altitude, the colder the temperature. The rate of emission of thermal infrared decreases with temperature. By forcing the average photon escaping to space to be emitted from a higher altitude, GHGs slow down the rate at which thermal IR reaches space. In an isothermal atmosphere, there would be no GHE.

The climate models used by scientists do not assume that thermal IR is trapped by GHGs. They assume that the emission of thermal IR by GHGs in the atmosphere depends on the local temperature and not on how many photons have been “trapped” and not thermalized. That is why they correctly predict the spectrum and intensity of thermal IR reaching the ground and space – as shown by the figures shown by our host.

2. I was looking through old papers, Manabe I think, and came upon the seasonal response of relative humidity. Not just absolute humidity, but relative humidity was higher during summer than winter. This is an imperfect, but real world example of water vapor increasing with temperature, and one that is stronger than just constant relative humidity.

Of course, increased water vapor is probably correlated with decreased extremes of temperature while at the same time increasing the mean. And the missing hot spot raises some interesting distribution effects. But WV Feedback may larger than idealized.

• This paper compares different reanalyses wrt RH, but also depicts the seasonal cycle:

Some of that change is due to dynamics, of course.
But for the tropics, anyway, the correlation of of humidity with temperature is even greater than constant RH.

3. SoD,

It’s still not at all clear to me that constant relative humidity is truly an emergent property in a climate model rather than the result of tuning the evaporation and precipitation parameters. After all, the rate of increase of precipitation with temperature varies significantly across the population of models. Yet, AFAIK, they all have relatively constant RH with temperature.

• DeWitt,

As a subsidiary point to the main point of this article, “Is constant relative humidity effectively programmed in by tuning of evaporation and precipitation parameters by climate modelers“?

(Note for beginners – this is a completely different proposition from the title of this article).

If you ask me – can I prove the opposite to your suggestion? I don’t know. At the very least, it would take some work.

I have accepted a point of view from luminaries in the field like Isaac Held and many others via many papers stating it. And I have never seen any evidence for your proposition.

You should make your case in more detail.

• After all, the rate of increase of precipitation with temperature varies significantly across the population of models.

And that was a key component of the so-called Iris Effect – precipitation efficiency.

Increased precipitation efficiency tends to dehumidify the upper atmosphere.
Decreased precipitation efficiency tends to humidify the upper atmosphere.

And the profile of changes in humidity with height change the radiative forcing.

Greater increase of humidity lower in the atmosphere,
with lesser increase of humidity higher in the atmosphere,
tends to increase the outgoing IR, reducing somewhat, the RF.

Lesser increase of humidity lower in the atmosphere,
with greater increase of humidity higher in the atmosphere,
tends to decrease the outgoing IR, thus causing more feedback.

• It’s still not at all clear to me that constant relative humidity is truly an emergent property in a climate model rather than the result of tuning the evaporation and precipitation parameters.

Humidity should be an emergent property of the evaporation and precipitation parameters.

So.. can you clarify what you’re saying?
Are you just saying that the evaporation and precipitation parameters are incorrect?

• Windchaser,

I’m assuming that precipitation rate in the models has to be a function of relative humidity, among other things. A choice of evaporation rate parameters that gives a small change in evaporation rate with temperature would force the relative humidity to be constant. That’s because evaporation and precipitation has to balance globally. To me, that makes constant RH a modeling choice, not an emergent property, even if it’s not programmed directly.

• DeWitt,

1. The relative humidity in a region is dependent on:

– the horizontal winds (carrying water vapor)
– the evaporation rate at the surface and the vertical convection rate
– the condensation rate

2. The vertical convection is dependent on:

– the environmental lapse rate (=[actual temperature in a grid cell – actual surface temperature]/altitude
– the adiabatic lapse rate, i.e. the buoyancy neutral value of the lapse rate

3. The actual temperature in a grid cell is dependent on:

– radiative heating/cooling which is dependent upon the GHGs
– convective heat transfer (item 2 above)
– horizontal winds (carrying heat)

Some of the above items are affected by clouds. I probably missed a few items, but this is just off the top of my head.

Can you explain how modelers are consciously or sub-consciously able to make relative humidity constant by their modeling choices.

• SoD,

You’ve left out precipitation. That’s critical to my conjecture. Evaporation must equal precipitation globally. Evaporation increases humidity, precipitation decreases it. If precipitation is high, the RH goes down, which increases evaporation. If precipitation is low, RH increases and evaporation decreases. So the rate of increase of precipitation with temperature will, I think, control humidity. The other humidity variables will adjust.

IIRC, models do a lousy job on precipitation in both amount and location.

• SoD,

See also the quote from Held in vtg’s post below. I read that as confirming my thought that the model hydrological cycle determines the constancy of RH. Given the size of the grid cells, I still think that may be a modeling choice in the tuning process, not necessarily an emergent property. Note my use of ‘may’.

4. SOD, The problem here is that precipitation is an ill posed problem driven as it is by convection. The parameterizations of convection are very questionable. Isaac Held has an old blog post on this showing that the size of the computational domain has a huge effect on the aggregation of convective cells. You rightly ridicule simplistic characterization of GDM’s but sometimes those simplistic formulations have an important grain of truth. Further convection in the tropics is critical to the existence of the “hot spot” which even consensus scientists are starting to admit is not really there. Real Climate has a graph showing the divergence between the data and modeling results. Held also in a recent blog post says that if this modeling of the tropical troposphere is wrong GCM’s will need major rework even though he doesn’t think the evidence is conclusive.

• dpy6629,

This water vapour feedback is simply written into climate models as parameters. It is not derived from any kind simulation of first principles in the General Circulation Model runs (GCMs)..

Answer – No. The feedback does emerge from the properties of the GCM.

It’s a very basic point.

• Yes SOD, your point is entirely correct. I’m just saying that there may be more circuitous ways in which parameterizations affect the modeled water vapor feedback.

• Further convection in the tropics is critical to the existence of the “hot spot” which even consensus scientists are starting to admit is not really there.

And this is loaded with implications.

The lack of “hot spot”, at least for the satellite era, means also a reduction of water vapor feedback and a lack of “lapse rate feedback” for the hot spot.

And the vertical profile of GHGs matter.

Imagine all the water vapor of the atmosphere was contained in the lowest 1 meter of the atmosphere – the atmosphere would emit more to space.

Imagine all the water vapor of the atmosphere was contained in the highest 1 meter of the troposphere – the atmosphere would emit less to space.

WV feedback for a “hot spot” implies a greater percentage increase of WV aloft.
Reality indicates a greater percentage increase of WV nearer the surface.

• And this is loaded with implications.

I’m not sure quite what the implications are (for a reduction in water vapor feedback in the upper troposphere), in terms of climate sensitivity. On the one hand, you lose the positive radiative feedback this provides. On the other hand, there’s a negative feedback when water vapor transports heat high into the atmosphere where heat can be more easily lost to space.

I’ve heard that the lack of a “hot spot” is basically a wash; these two feedbacks cancel out when the tropospheric warming is greater than the surface warming. But that’s just what I’ve heard — nor do I know of studies that actually examine these effects of potentially-reduced convection into the high troposphere.

• Windchaser wrote: “I’m not sure quite what the implications are (for a reduction in water vapor feedback in the upper troposphere), in terms of climate sensitivity. On the one hand, you lose the positive radiative feedback this provides. On the other hand, there’s a negative feedback when water vapor transports heat high into the atmosphere where heat can be more easily lost to space.”

Climate models show a rather large range of values for the water vapor feedback. The ones with a large water vapor feedback have a strong negative lapse rate feedback while the models with weaker water vapor feedback have a weaker lapse rate feedback. That is, I think, consistent with Winchaser’s reasoning. In all the models, the sum of the water vapor and lapse rate feedbacks is very nearly the same. That suggests that clear sky effects are not sensitive to the variations between models. But clouds are a different matter.

It seems to me that the fact that models give significantly differing results for the water vapor feedback is pretty strong evidence that constant RH is not built in to the models.

Windchaser wrote: “I’ve heard that the lack of a “hot spot” is basically a wash; these two feedbacks cancel out when the tropospheric warming is greater than the surface warming.”

They only partially cancel. I think that models with strong water vapor and lapse rate feedbacks give a strong hot spot and ones in which those feedbacks are weaker give a weaker hot spot.

• Mike M.,

It seems to me that the fact that models give significantly differing results for the water vapor feedback is pretty strong evidence that constant RH is not built in to the models.

I think you may be conflating water vapor and cloud feedback when you say that the models have significantly differing results for water vapor feedback. Of course since the models have different global average absolute temperatures, there’s going to be a significant difference in specific humidity as well,

• DeWitt wrote: “I think you may be conflating water vapor and cloud feedback when you say that the models have significantly differing results for water vapor feedback.”

No, CMIP5 models have a range of 1.4 to 1.9 W/m^2/C for the water vapor feedback and CMIP had a range of about 1.5 to 2.3. Not nearly as bad as the cloud feedback, but still a pretty wide range (although not so large as I remembered).

• Mike M.,

Those ranges are not unreasonable. MODTRAN with the 1976 US Standard Atmosphere has a difference at the tropopause of ~1.2W/m² between constant vapor pressure and constant RH for a 1 degree change in surface temperature. The specific humidity of that atmosphere is pretty low.

• DeWitt Payne wrote: “Those ranges are not unreasonable. MODTRAN with the 1976 US Standard Atmosphere has a difference at the tropopause of ~1.2W/m² between constant vapor pressure and constant RH for a 1 degree change in surface temperature.”

Thanks, that provides some useful context. If I understand you correctly, that means that the CMIP3 range (0.8 W/m^2/K) is equal to 2/3 of the difference between constant RH and constant partial pressure, while the CMIP5 range (0.5 W/m^2/K) is 40% of the difference. So the models do not all give constant RH.

Almost all the water vapor in the atmosphere is in the boundary layer. I think the models do give very nearly constant RH for the boundary layer. But the feedback is due to water vapor in the upper troposphere where it seems the RH varies between models.

5. “just show where this water vapor feedback is written in”
It isn’t, of course. And one reason is, there is nowhere to write it. The structure of the model is as SoD describes, and is purely local. It is very hard to apply global constraints such as global feedback equations. Where done, it would have to be by some sort of tuning. It can’t be built into the dynamics of an explicit time-stepping system.

• There are certainly excessive and dubious claims in the blogosphere on both sides of the fence. One of them is that precipitations should augment as humidity augments. Some even seem to contend the silly belief that it might increase as humidity itself from Clapeyron equilibrium, about 7 % per K. Yet, besides the fact that there is no causal relationship between water flux and water reservoir or stock in steady state, every % of increase in precipitations implies a relevant increase in latent heat transport upwards of near 1 W/m2. So already a modest 3% increase would essentially cancel the whole radiative forcing brought about by anthropic CO2.

• Yet, besides the fact that there is no causal relationship between water flux and water reservoir or stock in steady state, every % of increase in precipitations implies a relevant increase in latent heat transport upwards of near 1 W/m2. So already a modest 3% increase would essentially cancel the whole radiative forcing brought about by anthropic CO2.

gammacrux, can you explain your reasoning here?

Yeah, I get how an increase in precipitation means an increase in latent heat transport. But the jump from latent heat transport to “this cancels out CO2 forcing” is fuzzy.

Latent heat from condensation doesn’t just leave the atmosphere. The heat is moved from the point of evaporation to the point of condensation. So before you can calculate the temperature effects of this, you’d need to know where the condensation occurs, yah?

That heat may have an easier time leaving the system, being at a higher altitude than it started at, but it still has to get out of the atmosphere. So I’m pretty sure you can’t just say “1 W of latent heat transport offsets 1 W of CO2 forcing”.

• Windchaser,

I’m pretty sure you can’t just say “1 W of latent heat transport offsets 1 W of CO2 forcing”.

Without the current 80W/m² of latent heat transfer (TFK, 2009), the surface temperature would have to increase from 288K to 302K to a first approximation. One would have to know how much the sensible heat transfer would increase as well as how much of the increased surface radiation would escape to space directly to get a better approximation. The lapse rate without latent heat transfer would probably increase, resulting in, I think, less radiation from the atmosphere.

Running the calculation on MODTRAN using the 1976 US Standard Atmosphere with constant RH, it’s not 1 for 1. It’s more like 2W/m² at the surface for 1W/m² at the TOA.

• Windchaser

I agree that latent heat is generally not simply released at the altitude of emission to space. Yet don’t deep convection in tropical zone or in a hurricane essentially do it?

Moreover enhanced precipitations and latent heat transport implies more energy that flows into the convection “thermal engine” and thus more convection and sensible heat transport, too.

And even if 2 W/m2 at surface offset only 1 W/m2 at top of atmosphere as suggested by DeWitt Payne It seems to me that this puts a strong constraint on the possible increase in precipitations as a result of anthropic GHGs emission.

6. Many subjects that become politicized seem to end up being highly polarized.

I’ll use Stephen Pinker’s notation (he is from another discipline of cognitive psychology) of East and West to avoid familiar notations.

East – Climate Models ————————————- West – Climate Models
are so good we ———————————————- outputs are prescribed
can accept ————————————————— by climate modelers
their results with
little question

I don’t believe either. Rejecting one doesn’t mean accepting the other.
(Here’s hoping the formatting comes out ok).

7. Speaking of familiar notations, the pi-throwing in the math recalls Nicholas Vanserg’s essay “Mathmanship”. Pi-throwing is assigning a disparate meaning to a familiar symbol (such as π) to generate confusion.

8. SOD: Climate models actually predict an increases in relative humidity over the oceans with warming and a decrease in relative humidity over land.

Ocean: The rate of evaporation over water is proportional to “undersaturation” (100% – relative humidity) and wind speed. If relative humidity and wind speed remain constant, then the rate of evaporation would increase at the same rate as saturation vapor pressure (7%/K). A 7% increase in latent heat flux (80 W/m2) is 5.6 W/m2/K. An ECS of 3 K/doubling implies an increase in net radiative flux to space of 1.2 W/m2/K and 1.5 K/doubling an increase of 2.5 W/m2/K. At steady state, the increase in heat flux from the surface to the atmosphere must be the same as from the atmosphere to space, so climate models must have some mechanism for suppressing a 7%/K increase in evaporation and precipitation. Instead climate models predict only a 2%/K increase in evaporation/precipitation, which occurs mostly because relative humidity over the ocean rises from 80% to 81%, which is a 5% decrease in “undersaturation”

Land: Over land, a 2%/K increase in precipitation can’t keep up with a 7%/K increase in saturation vapor pressure. Relative humidity must decrease.

Since climate models predict that the rate of increase in precipitation with temperature with warming (about 2%/K) is much less than the increase in saturation vapor pressure (7%/K)

Isaac Held discusses relative humidity over the ocean at the link below:

https://www.gfdl.noaa.gov/blog_held/47-relative-humidity-over-the-oceans/

You discussed physics controlling evaporation and the 2%/K predicted increase in evaporation/precipitation at the links below:

9. As Isaac Held was mentioned in the comments, on the subject of the physical constraints and modelling of relative humidity (his subsequent post is also relevant)

https://www.gfdl.noaa.gov/blog_held/47-relative-humidity-over-the-oceans/

Models very robustly maintain more or less constant relative humidity in these lower tropospheric layers over the oceans as they warm, basically due to the constraint imposed by the energy balance of the troposphere on the strength of the hydrological cycle, and the tight coupling between the latter and the low level relative humidity over the oceans.

• VTG: I look at this issue from a different perspective from Isaac. I don’t presume models are correct.

Let’s imagine a K-T energy balance diagram, except that we will label it with the CHANGE in energy flux per degK of warming at equilibrium (W/m2/K).

An ECS of 3 means a 3.7 W/m2 forcing is compensated for at equilibrium by 3 K warming – meaning a climate feedback parameter of only 1.2 W/m2/K. This means that increased OLR from warming plus increased reflection of SWR can only total 1.2 W/m2/K. Of course, reflection of SWR (albedo) can decrease. A 1 W/m2/K decrease is a 1%/K decrease.

A variety of reasons suggest that the increase in net radiative cooling of the surface is near 0 W/m2/K. Upward emission from the ground increases by 4oT^3 or 5.4 W/m2/K. If DLR came from a black body, it would be 277 K and would increase 4.8 W/m2/K with constant lapse rate. Since its emissivity is less than 1, the increase must be greater 4.8 W/m2/K. With increasing CO2 and absolute humidity, DLR must originate from a little closer to the surface where it is a little warmer. The instantaneous change in DLR for 2XCO2 is 0.8 W/m2. One can also explore this issue with MODTRAN.

If evaporation/precipitation increased with saturation vapor pressure, that would be 7%/K or 5.6 W/m2/K. An ECS of 3 would then require a massive and improbably decrease in albedo, of about 4 W/m2/K or 4%/K. Instead, climate models predict only a 2%/K increase in precipitation.

However, the rate of evaporation is proportional to undersaturation (and wind speed), so it should rises 7%/K – unless relative humidity increases. This happens because the overturning of the atmosphere slows: Relative humidity over the oceans is less than 100% because of atmospheric overturning, so when it slows, relative humidity rises 1% (which is a 5% decrease in undersaturation and evaporation rate).

One take-home lesson from this long story is that high climate sensitivity depends on a slowing of atmospheric overturning and the hydrologic cycle (2%/K rather than 7%/K).

A second is that the change in saturation vapor pressure (7%/K) is associated with massive amount of energy (5.5 W/m2/K). When Held says models “very robustly maintain more or less constant relative humidity”, there are small changes in relative humidity, but they represent large amounts of energy. The physics of the situation doesn’t allow large changes, unless you wish to call a 5% reduction in undersaturation a “large change”.

• A third take-home lesson: “If you don’t have confidence in the ability of climate models to correctly model the slowdown in atmospheric overturning from rising GHGs, then you shouldn’t have confidence in their ECS. The two phenomena are intimately interconnected.

” This water vapour feedback is simply written into climate models as parameters. It is not derived from any kind of simulation of first principles in the General Circulation Model runs (GCMs)..”

SOD “Is constant relative humidity effectively programmed in by tuning of evaporation and precipitation parameters by climate modelers“?

DW “It’s still not at all clear to me that constant relative humidity is truly an emergent property in a climate model rather than the result of tuning the evaporation and precipitation parameters.”

SOD ” water vapor concentrations increase in a warmer world, the climatic effects of the other greenhouse gases will be amplified. Models of the Earth’s climate indicate that this is an important positive feedback that increases the sensitivity of surface temperatures to carbon dioxide by nearly a factor of two when considered in isolation from other feedbacks, and possibly by as much as a factor of three or more when interactions with other feedbacks are considered.”

The question of emergence of feedbacks, feedback amplification,CRH above, climate sensitivity from models is of great importance to all of us.
The quote initially appears flawed. Climate models have data and assumptions on how to use the data written into them, hopefully from first principles. Hopefully it is “simply”, as in accurately, written.Whether you argue it is from these first principles or from the assumptions that are made on how to use these first principles does not change the fact that one can honestly say the results are written into the models.
How one wishes to judge the output or to make claims on the assumptions is another matter.

– Nick Stokes touches on the sore point eloquently
“It is very hard to apply global constraints such as global feedback equations. Where done, it would have to be by some sort of tuning.”

We all agree that many climate models are tuned, I hope.
The tuning involves, must involve changes to the assumptions used as it cannot apply to the first principals.
[I stand to be corrected on this because I read somewhere of one case where wrong first principles were used precisely because they gave a right result].
Hence the factors which give the emergent CRH are amenable to manipulation.

In a post at ATTP I was forced to look up the effect of water vapour on the temperature of the earth. DeWitt [DW] touched on the effect of latent heat but what I found was a comment on the albedo effect of water vapour.
If water vapour did not reflect as much SW as it does the earth would be a lot hotter than it is for the GHG effect of water.
In other words increasing water vapour has a negative feedback effect that is very substantial and at odds with your quote of 2-3 times doubling.
This was not in dispute in early IPCC versions but is now considered wrong. Magically at 14.5 C water vapour which up to that temperature has a known negative feedback turns into a positive feedback, dare I say saving the models?

What I am saying is that assumptions are made on many water based parts of your cells. If the expectation of a doubling is made on first principles it must trigger over into all subsequent assumptions.
You may be quite at ease with this as the principles have been tested and found accurate. Or you might question why, with such accuracy, the models need a lot more tuning than expected.

• “In other words increasing water vapour has a negative feedback effect that is very substantial and at odds with your quote of 2-3 times doubling.”

Something tells me Angech is confused again, and mixing up stuff.

Water vapor does not equal clouds.

• If water vapour did not reflect as much SW as it does the earth would be a lot hotter than it is for the GHG effect of water.

Water vapor reflects and absorbs very little short-wave; it’s near-transparent.

Recall that “water vapor” describes H2O in its gaseous phase, not in liquid or solid phases. Clouds are made up of the liquid or solid H2O in suspension in the air, rather than the gaseous H2O. You get clouds only once H2O has precipitated out of water vapor.

In other words increasing water vapour has a negative feedback effect that is very substantial and at odds with your quote of 2-3 times doubling.
This was not in dispute in early IPCC versions but is now considered wrong.

[emphasis mine: citation needed]

Recall that water vapor has a very large direct effect as a radiative feedback. It is a strong greenhouse gas. There are also secondary feedbacks to having water vapor in the atmosphere: the convective transport of latent heat higher in the atmosphere, and clouds, which have both positive and negative feedbacks.

You need to provide a citation for your claim that the total of these water vapor-related feedbacks is negative.

• “water vapor-related feedbacks is negative”

It’s somewhat ironic, but if there are feedbacks which would reduce or even completely negate warming, they would occur through some change in the distribution of clouds,water vapor, temperature profile, etc.

So, if there were no warming, it would be because of climate change.

One observation of water vapor feedback is the seasonal cycle.

Coincident with this, the slope of the line associated with seasonal temperature response ( emissive temperature determined by OLR versus surface temperature ) is less than 1. This indicates a positive feedback of some kind. Of course, there are differences between the seasonal cycle and long term trend imposed by increased RF, but we can then consider why we might expect any of those differences to pertain in one case and not the other.

• In other words increasing water vapour has a negative feedback effect that is very substantial and at odds with your quote of 2-3 times doubling.

That’s incorrect. The albedo effect is, as others have pointed out, due to liquid water (droplets in clouds), not water vapour. Increasing water vapour with concomitant temperature increase does not necessarily increase liquid water.

This was not in dispute in early IPCC versions but is now considered wrong.

I think you just made that up. Happy to be proved wrong with a citation.

Magically at 14.5 C water vapour which up to that temperature has a known negative feedback turns into a positive feedback, dare I say saving the models?

You have comprehensively misunderstood the topic. Please provide a citation that water vapour feedback is claimed to change sign at 14.5 degrees.

11. SOD quotes: “This water vapour feedback is simply written into climate models as parameters”.

A better way to approach this question is to ask if climate models CAN be tuned to produce a water vapor feedback of about +2.2 W/m2/K or perhaps so they produce a combined WV+LR feedback of +1.1 W/m2/K, which is equivalent to Planck+WV+LR feedback of -2.1 W/m2/K. It is well-known from the seasonal cycle that feedback through clear skies is about -2.1 W/m2/K. All climate models show a combined WV+LR feedback of about +1.1 W/m2/K.

It is almost certainly possible that parameters could be tuned to match this observation while models are being tuned to agree with other observables. When modelers are forced to disclose their tuning strategy for CMIP6, then we should know if they are tuning to produce +1.1 W/m2/K. Until then, IMO we don’t really know.

12. “GATOR-GCMOM is“ the first fully-coupled online model in the history that accounts for all major feedbacks among major atmospheric processes based on first principles; and despite hundreds of processes in it still not in any other mode”

From Professor Jacobson, who should know. Zhang’s
2008 Atmospheric Physics and Chemistry Journal comprehensive review quoted. The first in 2008?

13. Marco
” Water vapor does not equal clouds.”
Windchaser
“Water vapor reflects and absorbs very little short-wave; it’s near-transparent.
Recall that “water vapor” describes H2O in its gaseous phase, not in liquid or solid phases.”

So very true and so true of the problem with pedants.

I can either write 50 lines with all the caveats or write a simple expression which everyone here should understand.
Water vapour was used as a generic term for all the forms of the GHG H20 in the atmosphere

What argument will you be left with now?

“Rising global temperatures are expected to cause greater evaporation of water vapor into the atmosphere, primarily from the oceans. On one hand, we know that water vapor is a powerful greenhouse gas, so an increase in water vapor might be expected to produce yet more warming through an enhanced greenhouse effect. This warming should further enhance evaporation, producing more water vapor, and leading to a “vicious cycle” (or “positive feedback loop”) of more and more warming… and eventually to a “runaway greenhouse effect”.

On the other hand, more water vapor in the air is likely to cause more clouds to form. The presence of clouds dramatically increases Earth’s overall albedo, reflecting a lot of the incoming sunlight back into space. Increased cloudiness would be expected to further reduce the amount of sunlight reaching our planet’s surface, thus providing a net cooling effect. Thus an increase in water vapor, and hence cloudiness, might actually serve as a “self correcting” mechanism (or “negative feedback loop”) that would “put the brakes on” global warming; or possibly induce a period of “global cooling”.

• Angech, that cloudiness is nice and all that, but…even there you are just scratching the surface. It is not just a matter of more clouds = increasing albedo – it is important WHERE those clouds are, both in terms of altitude and latitude, and what type of clouds. For example, Lindzen’s iris hypothesis, which suggested a negative feedback of clouds, was actually related to a *reduction* in cirrus clouds, thereby allowing more IR radiation to escape.

14. Windchaser
“You need to provide a citation for your claim that the total of these water vapor-related feedbacks is negative.”

The quote above fleshing out your argument and the half you conveniently ignored was from the “National Earth Science Teachers Association”

More to the point do you really not know that water vapour [the cloud component thereof] is the most important cause of keeping our temperature down from the horrendous amount of GHG effect it has?

If a surface area has an albedo of 10 percent and clouds an albedo of 90 percent then with 70 percent cloudiness the albedo for the area is 0.7(0.9)+0.3(0.1)=0.66=66 percent. This means that 34 percent of the Sun’s radiation for the area is not reflected back into space. If a climate model computes the cloudiness to be 40 percent then the albedo is 0.4(0.9)+(0.6)(0.1)=0.42=42 percent. This means the model would be using a solar input of energy to the area that was 42/66=0.636 of the actual, or roughly 64 percent. It is at first hard to see how the model could have an accurate computation of the temperature if the energy input was so far off. The answer as previously noted is that the temperature depends upon the fourth root of the energy input. The fourth root of 0.636 is about 0.893 or roughly 0.9, so the error in temperature would be only about 10 percent. Thus an error of only about 10 percent in temperature corresponds to an error of 34.4 percent in energy flow.

I will find the quotes from the old ATTP discussion and get back to you.

• “Thus an error of only about 10 percent in temperature”

Errr….on a Kelvin scale that 10% is not “only”, but essentially THREE degrees.

• Errr…oops! Make that THIRTY degrees.

• Marco,

An increase in planetary albedo of one percentage point, from 0.30 to 0.31, would reduce insolation by 3.2W/m². Climate models do clouds badly. I take any prediction of changes in cloud cover with temperature from models with about 50 lbs of salt.

15. SOD can confirm. No water, no GHG effect of water and no albedo effect from no clouds. As the water vapour increases the clouds form causing a marked albedo effect.This effect cancels out much of the rise that water vapour causes and must be taken into context apart from the straight water vapour effect itself, which does not seem to be the case.
You cannot have one without the other generally.

• angech wrote: “As the water vapour increases the clouds form causing a marked albedo effect.This effect cancels out much of the rise that water vapour causes and must be taken into context apart from the straight water vapour effect itself, which does not seem to be the case.”

There are three big problems with that statement. First, more water vapor does not automatically mean more clouds. That is not even particularly plausible if the increase in water vapor is due to increased temperature with constant RH.

Second, the effect of more clouds need not be more cooling. Clouds not only reflect short wave radiation, they absorb long wave radiation and contribute to the greenhouse effect. The relative importance of the cooling and warming effects depends on things like cloud location, density, and altitude.

Third, changes in clouds are most certainly taken into account in the models. Badly, I suspect, but definitely not ignored. If a model give high climate sensitivity, it is due to clouds.

• So, I like setting up little experiments for myself.
And along the lines of seasonal response above, would this then predict that for a given hemisphere, summers are cloudier than winters?

A simple examination of albedo, surface temperature, cloud cover ( at different elevations ) might elucidate.

• Arctic Sea ice blog short term graphs has a beautiful world humidity graph showing the concentrations on an active map of the world. You could see the hurricanes that hit America and Mexico forming a week or two beforehand. Most water vapour is over the tropical oceans and very dry and cold over the poles.
Having lived in Darwin I can vouch for the cloudiness in summer, 60 inches of rain and the clear skies in “winter”, still 28 C , but no rain for 6 months, the dry season.
Question is the albedo effect of clouds double the daily stated rate as they only do this during the day for sunlight or is albedo only the sunlight effect (day albedo?) with a minuscule amount for stars and moon?

• angech,

Which season in the Northern Hemisphere mid-latitudes is warmest?

In which season in the Northern Hemisphere mid-latitudes is water vapor at its highest concentration?

Which season in the Northern Hemisphere mid-latitudes has the least cloud cover?

Hint: Same answer for all three.

• paulski0
Initial pass, TE raised the issue.
However just to test your very good understanding and improve mine I will bite.

Looked these facts up
“When the convection cells are moved north and south as the seasons change, climate temperatures follow, but with a time lag due to the time it takes for the sun to heat up the land or ocean. This is the reason why many Northern Hemisphere mid-and high-latitude areas have their warmest weather in August and their coldest in February, rather than at the solstices in June or December.
Approximately 93% of water evaporated to the atmosphere is evaporated from the oceans but only 71% of the total precipitation (rain and snow) falls on the oceans.
Atmospheric upwelling and downwelling areas are regions where winds are light. At upwelling areas clouds and rain are abundant, whereas at downwelling areas there is little or no rainfall.
Atmospheric upwelling occurs at the equator and between the Ferrel and polar cells. Atmospheric downwelling occurs between the Hadley cells and Ferrel cells at around 30o latitude and at the poles.
Because ocean water has a high heat capacity compared to land, convection cells migrate farther north and south seasonally over the continents than they do over the oceans”

Taking a stab at your question[s]
The season you seem to be suggesting is Autumn rather than Summer because some land areas have their warmest weather in August and are prone to an atmospheric circulation effect which modifies the atmospheric behaviour re cloud formation and precipitation

You may still have your humorous facts wrong.
Unlikely but…
Humorous subterfuge being to find examples of weather behaviour which go against the general physical behaviour of water vapour and clouds [formed of ice particles etc] in a local area.
The fact that Coriolis forces are an extra effect in these areas seems to delight you.
But it might sadden VTG to learn of extra effects, [This means that atmospheric water vapour and temperature are tightly coupled, as the absolute humidity at the ocean/atmosphere interface is dependent on the temperature.
So water vapour can only increase if temperature increases.]

Please give your source for all 3 answers and note, the Northern Hemisphere has more land than the south but most of the action with water vapour and clouds occur over sea.
[Earlier maximum temp and less Northwards Hadley cell movement].
It has to be the entire middle latitudes [30-60] including oceans.
I do not mind if you are right.
Waiting

Would you also care to posit the season, cloud cover and water vapour concentration for the upper and lower latitudes to round out our education or do they behave as TE said?

• paulskio?
Are you right?
Can you show it?
Which season in the Northern Hemisphere mid-latitudes has the least cloud cover?
Winter would still seem to be correct for the entire mid latitude NH, yes?

16. Mike M,

“There are three big problems with that statement. First, more water vapor does not automatically mean more clouds. That is not even particularly plausible if the increase in water vapor is due to increased temperature with constant RH.”

Look, even if you are right you are wrong. Cloud formation, water cloud formation is a complex business involving a lot more than just water vapour.
There are temperature, height, aerosols and other factors involved in forming clouds.
Let us just agree on earth as the site, now as the time, and a global temperature range somewhere sensible. Ie not 100 C above or below where we are now. More water vapour means more clouds is a simple statement of fact. No water, no clouds we might be able to agree on. Some water some clouds possible (even if it is water in the form of ice crystals or whatever).
In this sensible context it is very hard not to argue that more water vapour in the air means more clouds.
You can pick unicorns, you can choose over the top of the poles or the Sahara desert but basically, common sense discussion wise, more after vapour does mean more clouds.

“Second, the effect of more clouds need not be more cooling.”

I have found my money quote courtesy of Victor Venema.
“Clearing clouds of uncertainty” by Mark Zelinka et al Nature Climate change see ATTP blog comment Victor Venema says: October 7, 2017 at 11:30 pm
“Averaged globally and annually clouds cause -18 Watts/m2 of cooling compared to a cloud free earth.”
The net planetary cooling if clouds is 5 times the warming of a doubling of CO2

“Clouds not only reflect short wave radiation, they absorb long wave radiation and contribute to the greenhouse effect. The relative importance of the cooling and warming effects depends on things like cloud location, density, and altitude.”
Is misleading and irrelevant, the relative importance of the cooling effect far outweighs the minor details mentioned.

“Third, changes in clouds are most certainly taken into account in the models.
Badly, I suspect, but definitely not ignored.”
Great comment.

“If a model give high climate sensitivity, it is due to clouds.”

Despite SOD and Lucia and Judith all describing CS as an emergent property, it is purely a choice of terms .
If A+B=X
Then the output is absolutely dependent on the way the input is handled.
CS and water vapour feedback may be difficult to predict and seem to emerge from a model but they are absolutely contingengent on the setup and parameters, first principles or more likely not, that are put into the makeup of the models.

I would be interested in your reflections on these statements and hope you can read the article re an often misunderstood property of clouds (water if not vapour)

17. In this sensible context it is very hard not to argue that more water vapour in the air means more clouds.

As a thought experiment on a planet without liquid water at the surface, this might be reasonable.

However, the earth’s surface has an effectively infinite source of water vapour – the oceans.

This means that atmospheric water vapour and temperature are tightly coupled, as the absolute humidity at the ocean/atmosphere interface is dependent on the temperature.

So water vapour can only increase if temperature increases. That means that cloudiness, to a first approximation, might reasonably be expected to be independent of temperature, the increasing temperature cancelling the effect of increased absolute humidity. Relative humidity might, to a first approximation, be expected to be constant with changing temperature.

Of course, it’s more complex than that, but your assertion is clearly not supportable, for an earth – like planet.

• vtg,

That all sounds good until you look at model results. Values of climate sensitivity to doubling CO2 greater than 3K are caused by positive cloud feedback. Increased specific humidity alone won’t get you there. Clouds do change with temperature in the models. But there is low confidence that the models do clouds even close to correctly.

• I agree with you entirely DeWitt. I was taking issue with the “more water vapour = more clouds = negative feedback” point Angech was making.

I’m not making any strongly constrained claim for what cloud feedback is. I’ve not seen anyone convincingly do so – it seems pretty uncertain with a median of slight positive but wide uncertainty.

According to AR5, section 7.26,

…a 90% (very likely) range of −0.2 to +2.0 W m–2 °C–1, with a 17% probability of a negative feedback.

Do you have a different view?

• vtg,

I think that range may be biased high by model results. But that’s just a feeling without any quantitative support. I’m in the lukewarmer camp. I think it’s unlikely that the equilibrium climate sensitivity to doubling CO2, angech’s CS(?), is lower than 1.5K or higher than 3K.

• I think it’s unlikely that the equilibrium climate sensitivity to doubling CO2, angech’s CS(?), is lower than 1.5K or higher than 3K.

That’s quite a striking claim – the 3K upper limit.

Do you have any specific reason for such an opinion beyond “just a feeling”?

Genuine question.

• VTG and DeWItt: The IPCC’s values for cloud feedback summarize a very complex story that has only come to light since the conflict with EBMs became apparent. In climate models, both LWR and SWR cloud feedback increase with time and that increase is largest in the equatorial Pacific. In other words, feedbacks vary with both TIME and REGION. One might say that a runaway GHE develops in the Pacific, but doesn’t tip the whole planet into a runaway GHE. Some models agree with EBMs up until the present in RCP 8.5 runs, with feedbacks consistent with an ECS around 1.5-2.0 and then develop stronger cloud feedbacks consistent with higher ECS in the future. In 4X runs, which are used to extrapolate ECS, the first 20 years of warming show a very different slope than the last 150 years because cloud feedback is changing. So the feedbacks that are going to make AGW catastrophic by the end of the century can’t be observed today.

https://doi.org/10.1175/JCLI-D-14-00545.1 (Andrews et al 2015)

Experiments with CO2 instantaneously quadrupled and then held constant are used to show that the relationship between the global-mean net heat input to the climate system and the global-mean surface air temperature change is nonlinear in phase 5 of the Coupled Model Intercomparison Project (CMIP5) atmosphere–ocean general circulation models (AOGCMs). The nonlinearity is shown to arise from a change in strength of climate feedbacks driven by an evolving pattern of surface warming. In 23 out of the 27 AOGCMs examined, the climate feedback parameter becomes significantly (95% confidence) less negative (i.e., the effective climate sensitivity increases) as time passes. Cloud feedback parameters show the largest changes. In the AOGCM mean, approximately 60% of the change in feedback parameter comes from the tropics (30°N–30°S). An important region involved is the tropical Pacific, where the surface warming intensifies in the east after a few decades. The dependence of climate feedbacks on an evolving pattern of surface warming is confirmed using the HadGEM2 and HadCM3 atmosphere GCMs (AGCMs). With monthly evolving sea surface temperatures and sea ice prescribed from its AOGCM counterpart, each AGCM reproduces the time-varying feedbacks, but when a fixed pattern of warming is prescribed the radiative response is linear with global temperature change or nearly so. It is also demonstrated that the regression and fixed-SST methods for evaluating effective radiative forcing are in principle different, because rapid SST adjustment when CO2 is changed can produce a pattern of surface temperature change with zero global mean but nonzero change in net radiation at the top of the atmosphere (~−0.5 W m−2 in HadCM3).

http://onlinelibrary.wiley.com/doi/10.1002/2016GL068406/abstract;jsessionid=D9B8CE954979902C28851B4F282C30BD.f03t03

We investigate the climate feedback parameter α (W m−2 K−1) during the historical period (since 1871) in experiments using the HadGEM2 and HadCM3 atmosphere general circulation models (AGCMs) with constant preindustrial atmospheric composition and time-dependent observational sea surface temperature (SST) and sea ice boundary conditions. In both AGCMs, for the historical period as a whole, the effective climate sensitivity is ∼2 K (α≃1.7 W m−2 K−1), and α shows substantial decadal variation caused by the patterns of SST change. Both models agree with the AGCMs of the latest Coupled Model Intercomparison Project in showing a considerably smaller effective climate sensitivity of ∼1.5 K (α = 2.3 ± 0.7 W m−2 K−1), given the time-dependent changes in sea surface conditions observed during 1979–2008, than the corresponding coupled atmosphere-ocean general circulation models (AOGCMs) give under constant quadrupled CO2 concentration. These findings help to relieve the apparent contradiction between the larger values of effective climate sensitivity diagnosed from AOGCMs and the smaller values inferred from historical climate change.

• Frank wrote: “In 4X runs, which are used to extrapolate ECS, the first 20 years of warming show a very different slope than the last 150 years because cloud feedback is changing.”

The non-linearity in the Andrews et al. paper does not kick in until warming exceeds 5 K. So I’d put that under the heading of “no worries”. From what they say in the “Summary and discussion” it looks like the cause is not direct cloud feedback. It is changes in clouds due to changes in sea surface temperature distributions which in turn are due to changes in the overturning circulation. Since the models do a bad job on similar phenomena, like the PDO and AMO, I’d say that result has low credibility.

• Thanks for the reply, Mike

Mike M wrote: “The non-linearity in the Andrews et al. paper does not kick in until warming exceeds 5 K. So I’d put that under the heading of “no worries”. From what they say in the “Summary and discussion” it looks like the cause is not direct cloud feedback. It is changes in clouds due to changes in sea surface temperature distributions which in turn are due to changes in the overturning circulation.”

Most of the Andrews paper is based on experiments with one model (HAD3) and you are correct to point out that non-linearity doesn’t kick in until warming exceeds 5 K. Most of the work assumes a breakpoint occurring at 20 years (1-20 vs 21-150 yrs). Figure 3 is based on the mean of CMIP5 models and there the effects kick in at +3 K, perhaps even +2.5 K. This would be relevant.

IIRC, ECS is calculated from 4X experiments while feedbacks are calculated from 1% pa or RCP8.5 experiments. If you add up those feedbacks, they are inconsistent with the ECS from 4X experiments. The high values the IPCC reports for ECS (from 4X) are result of non-linear cloud feedbacks.

4XCO2 experiments produce an absurd planet with an ocean mixed layer 5 K warmer than today with little time for heat to reach the deep ocean. It is hard to imagine how the THC would continue to function under these circumstances, but ECS – in theory – is independent of ocean heat uptake.

• Frank,

You wrote: “Most of the work assumes a breakpoint occurring at 20 years (1-20 vs 21-150 yrs).”

But is time or amount of warming the relevant factor? If it is time, then the same thing would happen with, say, 2XCO2. But I see no indication of that.

If you give a system a hard kick, you should expect different transient behavior than if you apply steady pressure. I see no reason to believe that a transient following a 4XCO2 change would have anything to do with reality.

You wrote: “Figure 3 is based on the mean of CMIP5 models and there the effects kick in at +3 K, perhaps even +2.5 K. This would be relevant.”

That is only relevant if such warming were to occur. I see virtually zero likelihood of that, at least not in this century.

You wrote: “If you add up those feedbacks, they are inconsistent with the ECS from 4X experiments.”

Yes, that is in AR5, Fig. 9.43(b). So far as I know, the problem is not just for 4XCO2; it occurs in all models, most of which are very near linear up to 4XCO2. I did not know that there was an accepted explanation.

You wrote: “The high values the IPCC reports for ECS (from 4X) are result of non-linear cloud feedbacks.”

What is the basis of your claim? Direct cloud feedbacks would produce a rapid response. A delayed response would surely be due to a change in ocean circulation triggering other changes, such as a change in cloud patterns. To believe the models on this point would require believing that get a correct result for a combination of two things (clouds and MOC) that they are known to do badly.

You wrote: “4XCO2 experiments produce an absurd planet with an ocean mixed layer 5 K warmer than today with little time for heat to reach the deep ocean…”

Yes, especially if you mean an abrupt 4XCO2 change. So I don’t see why the result would be very relevant.

You wrote: “… It is hard to imagine how the THC would continue to function under these circumstances, …”

The overturning circulation is driven by the wind, so it would surely continue. But the deep ocean can’t be colder than the coldest parts of the surface ocean, so if the Arctic and Antarctic were to warm enough, the pattern of the overturning could change drastically.

You wrote: “… but ECS – in theory – is independent of ocean heat uptake”.

But if the overturning circulation were to drastically change, equilibrium would take thousands of years to be established. The models are not run that long.

18. angech wrote: “In this sensible context it is very hard not to argue that more water vapour in the air means more clouds”.

Ignorant nonsense. Verytallguy has saved me the trouble of adding details.

angech wrote: “The net planetary cooling if clouds is 5 times the warming of a doubling of CO2 … the relative importance of the cooling effect far outweighs the minor details mentioned.”

The “minor details” amount to eight or nine times the warming due to doubled CO2. Yes, the cooling effect of clouds is even larger. More area covered by cloud, with the distribution otherwise the same, would lead to cooling. But shifts in the latitude, altitude, and density of clouds will shift the balance between warming and cooling. That makes the net change hard to predict. Modelers say the result is more warming. I say we don’t know. Angech feels free to invent the answer he wants.

angech wrote: “describing CS as an emergent property”.

Once again, angech chooses to use his own private language, without regard to whether it can be understood by others.

19. VTG and Mike M.

“I was taking issue with the “more water vapour = more clouds = negative feedback” point Angech was making.”

Why?
I gave a reference which you could read,
In a paper desperately trying to cobble together a positive feedback.

Let us try again.
No water the earth would resemble the moon in temperature.
Water vapour, the main GHG, adds temperature to the atmospheric layer and earth surface commensurate on its overall average percentage in the air, less the amount of energy reflected back into space by the albedo effect when it is in cloud form ( ice crystals, whatever).
The amount of energy reflected back is known
18Watts/m2 annually averaged.
Where is the problem with this known, proven, documented by real scientists, very large and actual negative water base negative feedback?

Remember no water vapour, no clouds no negative feedback
Increases with water vapour and clouds to the current amount at the current state of affairs on a real earth

• Unfortunately, the Zelinka paper is paywalled. Do you have a link to a free copy?

• angech wrote: “Water vapour, the main GHG, adds temperature to the atmospheric layer and earth surface commensurate on its overall average percentage in the air, less the amount of energy reflected back into space by the albedo effect when it is in cloud form ( ice crystals, whatever).
The amount of energy reflected back is known
18Watts/m2 annually averaged.”

Spectacularly wrong. The amount reflected by clouds is between 45 and 50 W/m2. Clouds absorb about 30 W/m^2. Liquid water is much better at absorbing IR than water vapor. Some clouds absorb a lot more than they reflect. Others reflect much more than they absorb. If you change the patterns of clouds in time and space, you change the balance of warming and cooling.

angech wrote: “Remember no water vapour, no clouds no negative feedback
Increases with water vapour and clouds to the current amount at the current state of affairs on a real earth”

The combined effect of water vapor and clouds is strong warming. So a naive result, such as angech seems enamored of, should be a net positive feedback.

• Mike M.,

Read Zelinka. 18 W/m² is the net global cooling effect of clouds.

Averaged globally and annually, clouds cause ~18W/m² of cooling relative to a hypothetical cloud-free Earth

• DeWitt,

That is what I said. Angech gave 18 as the total, not the net.

I could have been more careful with my quotation marks.

• Mike M.,

angech is not known for the precision of his writing. I would give him a pass on that.

20. ater vapour in the air means more clouds.

“As a thought experiment on a planet without liquid water at the surface, this might be reasonable.”

Water in the air means water at the surface except when the surface temp is over 100 degrees at normal earth atmospheric pressure.
I thought I made it clear in an earlier comment that we should be addressing realistic, real world conditions.
In fact I mentioned this very proviso.

“However, the earth’s surface has an effectively infinite source of water vapour – the oceans.”

No, the amount of water is finite and only accessible easily directly over the oceans, this means that some of the water vapour has to travel in with the temperature changes from day to night.
It is not infinitely easily accessible.

“This means that atmospheric water vapour and temperature are tightly coupled, as the absolute humidity at the ocean/atmosphere interface is dependent on the temperature.”

Hence you choose to ignore the 1/3rd of the earth surface called land and the gradients between it and the sea where there cannot , by lack of proximity to your infinite anmount of water, be a tight coupling.
The chaos this throws up, along with the fast changing day night temperatures, means that the coupling cannot give humidity directly dependent to temperature except over the oceans.

“So water vapour can only increase if temperature increases.”

No, atmospheric pressure changes etc

“That means that cloudiness, to a first approximation, might reasonably be expected to be independent of temperature”

I think paraphrasing the bit further down answers this,
Of course, it’s more complex than that, this assertion is clearly not supportable, for a real earth – like planet.
Which is what I think we should be discussing

• Angech,

Let’s go back to your original claims

Firstly

This was not in dispute in early IPCC versions but is now considered wrong.

Please provide a citation that demonstrates this.

Magically at 14.5 C water vapour which up to that temperature has a known negative feedback turns into a positive feedback, dare I say saving the models?

Please provide a citation that water vapour feedback is claimed to change sign at 14.5 degrees.

Once we understand where these beliefs come from, it might be possible to have a more productive conversation about them.

Thank you.

• verytallguy Angech,
Let’s go back to your original claims.

More than happy to do so and thanks.

Tihis is all very well known and old hat
Working Group I: The Scientific Basis
“The role of clouds in the climate system continues to challenge the modelling of climate (e.g., Chapter 7, Section 7.2.2). It is generally accepted that the net effect of clouds on the radiative balance of the planet is negative and has an average magnitude of about 10 to 20 Wm-2. This balance consists of a short-wave cooling (the albedo effect) of about 40 to 50 Wm-2 and a long-wave warming of about 30 Wm-2. Unfortunately, the size of the uncertainties in this budget is large when compared to the expected anthropogenic greenhouse forcing. Although we know that the overall net effect of clouds on the radiative balance is slightly negative, we do not know the sign of cloud feedback with respect to the increase of greenhouse gases, and it may vary with the region. In fact, the basic issue of the nature of the future cloud feedback is not clear. Will it remain negative? If the planet warms, then it is plausible that evaporation will increase, which probably implies that liquid water content will increase but the volume of clouds may not.”

This is where you came in I think with,
“So water vapour can only increase if temperature increases. That means that cloudiness, to a first approximation, might reasonably be expected to be independent of temperature, the increasing temperature cancelling the effect of increased absolute humidity. Relative humidity might, to a first approximation, be expected to be constant with changing temperature.”

On an if which contradicts the history of cloud formation and negative feedback up to this level of current global temperature [the basic issue of the nature of the future cloud feedback is not clear].
As we both agree the cloud feedback is negative, was zero when it started[no clouds] and has increased up until this current temperature.
In that case there is little need for concern.
To have a problem one needs a positive feedback starting from our current temperature.
Magically, as in we need to believe this to move to step 2, evaporation increases and more water vapour but now at 14.5 C [the current world temp approx] the volume of clouds may not.
Just like that.
Plausible?
To cherry pick a point and say at just this temp all the other processes that went on below this temp suddenly stop?
Well we do have boiling points, freezing points and now the magical adiabetic relative humidity Cloud puncturing point MARHCP point.
Not to put to fine a point on it.

I apologise for my upset leading to my derogatory attempt at humour but this was to be a sensible debating point.
There may be a MARHCP point under another name which really exists. Happy as I said to learn”

• Angech,

I have replied to your points separately, at the bottom of the thread.

21. Angech and others: I am mystified as to why no one seems to think about what I see as the elephant in the humidity room, ENERGETIC CONSTRAINTS that force the hydrologic cycle to slow with warming (decreased overturning). Usually such mysteries mean I am in left field or wrong, but probably not in this case.)

A climate sensitivity of 3.7 K/doubling means that the net flux (OLR + reflected SWR) across the TOA increases by only 1 W/m2 per K of surface warming or 1 W/m2/K. 1.8 K/doubling = 2 W/m2/K. 1.2 K/doubling (improbable?) = 3 W/m2/K.

Saturation vapor pressure (which exists immediately adjacent to liquid water) rises 7%/K or 5.6 W/m2/K. It is energetically impractical for water vapor to continue to be transported upward in the atmosphere and precipitate at the current rate. Too much heat would build up in the atmosphere.

5.6 W/m2/K is bigger than any other climate phenomena including Planck feedback (3.2 W/m2/K) and water vapor feedback (2.2 W/m2/K, assuming constant relative humidity).

Water vapor and clouds everywhere will be changed by slower overturning. Climate sensitivity depends on how much overturning slows.

When slowed overturning produces much less than a 7%/K increase in precipitation, it is trivially obvious that relative humidity over land must drop. (AOGCMs predict only about a 2%/K increase in precipitation.)

The rate of evaporation over oceans depends on undersaturation (1 – RH) and wind speed. When slowed overturning produces much less than a 7%/K increase in precipitation, then one or both of these must fall – undersaturation according to models.

• Frank,

You wrote: “I am mystified as to why no one seems to think about what I see as the elephant in the humidity room, ENERGETIC CONSTRAINTS that force the hydrologic cycle to slow with warming (decreased overturning).”

I am unclear as to just what you mean.

You wrote: “Saturation vapor pressure (which exists immediately adjacent to liquid water) rises 7%/K or 5.6 W/m2/K.”

I guess the heat flux is the increase in upward latent heat transfer if the water cycle were to increase in direct proportion to vapor pressure.

You wrote: “It is energetically impractical for water vapor to continue to be transported upward in the atmosphere and precipitate at the current rate. Too much heat would build up in the atmosphere.”

Water vapor would be transported at a higher rate. There would be a reduced average lapse rate, resulting in slower upward transport of sensible heat and slower net radiative transfer in the lower atmosphere. I would expect the result to be greater than now, but less than implied by Clausius-Clapeyron. That seems to be what you say the models show.

You wrote: “5.6 W/m2/K is bigger than any other climate phenomena including Planck feedback (3.2 W/m2/K) and water vapor feedback (2.2 W/m2/K, assuming constant relative humidity).”

But if it is 2%, as you say later, that would be about 1.7 W/m^2/K. And that is an effect on near surface flux (over 500 W/m^2), not top of atmosphere flux (less than half as large).

You wrote: “When slowed overturning produces much less than a 7%/K increase in precipitation, it is trivially obvious that relative humidity over land must drop. (AOGCMs predict only about a 2%/K increase in precipitation.)”

I don’t see why. Weaker vertical transport would give higher RH in the marine boundary layer. There would be more transport out of that layer. I don’t see how that leads to lower RH elsewhere.

You wrote: “The rate of evaporation over oceans depends on undersaturation (1 – RH) and wind speed. When slowed overturning produces much less than a 7%/K increase in precipitation, then one or both of these must fall – undersaturation according to models.”

Yes, but I don’t see the point.

• Mike and I were discussing this flawed sentence I wrote: “It is energetically impractical for water vapor to continue to be transported upward in the atmosphere and precipitate at the current rate. Too much heat would build up in the atmosphere.”

Correction: It is energetically impractical for air to continue to be transported upward in the atmosphere at the current rate. That would carry 7%/K more moisture upward and release 7%/K more heat in the upper troposphere. Climate sensitivity limits how fast heat can escape through the TOA as the surface warms.

Frank wrote: “When slowed overturning produces much less than a 7%/K increase in precipitation, it is trivially obvious that relative humidity over land must drop. (AOGCMs predict only about a 2%/K increase in precipitation.)”

Mike replied: I don’t see why. Weaker vertical transport would give higher RH in the marine boundary layer. There would be more transport out of that layer. I don’t see how that leads to lower RH elsewhere.

If the supply of precipitation OVER LAND rises only 2%/K and saturation vapor pressure rises 7%/K, I think that relative humidity OVER LAND is likely to fall. However, if overturning over land slowed enough, relative humidity over land wouldn’t have to decrease (and could even increase as it will over the ocean due slower overturning.)

Mike comments: “but I don’t see the point.”

The point is that climate sensitivity and the slowing of overturning/hydrologic cycle/convection are firmly linked to each other by energetic considerations. No matter what climate scientists claim they know about feedbacks (Planck, WV, LR, LWR and SWR cloud, and surface), ECS depends on how much overturning slows with warming. If your AOGCM can’t correctly model this slowing, it can’t correctly predict ECS.

To put it differently, the difference between a 2%/K and 3%/K increase in precipitation is -0.8 W/m2/K. That is the difference between an ECS of 3 and 1.8 K/doubling. (:))

Mike wrote: “Weaker vertical transport would give higher RH in the marine boundary layer.”

And probably cause an increase in marine boundary layer clouds – which can reflect SWR without reducing OLR much and therefore are the most effective clouds at cooling. AOGCMs don’t do a good job with boundary layer clouds, probably because they don’t have the resolution to model to model this interface.

• Frank wrote: “It is energetically impractical for air to continue to be transported upward in the atmosphere at the current rate. That would carry 7%/K more moisture upward and release 7%/K more heat in the upper troposphere.”

I agree, with the minor quibble that you mean, of course, per degree.

Frank wrote: “Climate sensitivity limits how fast heat can escape through the TOA as the surface warms.”

Yes, although I think I’d say “radiative transfer” instead of “climate sensitivity”.

Frank wrote: “If the supply of precipitation OVER LAND rises only 2%/K and saturation vapor pressure rises 7%/K, I think that relative humidity OVER LAND is likely to fall. However, if overturning over land slowed enough, relative humidity over land wouldn’t have to decrease (and could even increase as it will over the ocean due slower overturning.)”

I agree. The relationship between averaged RH, clouds, and precipitation is not at all simple because changing transport rates are involved.

Frank wrote: “The point is that climate sensitivity and the slowing of overturning/hydrologic cycle/convection are firmly linked to each other by energetic considerations.”

That strikes me as an excellent explanation of the link between water vapor feedback and lapse rate feedback. Very nice.

Frank wrote: “No matter what climate scientists claim they know about feedbacks (Planck, WV, LR, LWR and SWR cloud, and surface), ECS depends on how much overturning slows with warming. If your AOGCM can’t correctly model this slowing, it can’t correctly predict ECS.”

That might overstate things. The models all give very nearly the same sum for water vapor plus lapse rate feedback. So it looks like they all get the big picture right even though the vary on the details. I suspect that is due to the tight linkage that Frank argues for. As long as the models are conserving things like energy and mass, they get that linkage right.

Frank wrote: “To put it differently, the difference between a 2%/K and 3%/K increase in precipitation is -0.8 W/m2/K. That is the difference between an ECS of 3 and 1.8 K/doubling.”

Careful. One is energy flux at the surface, but forcing is related to energy flux at top of atmosphere.

Mike wrote: “Weaker vertical transport would give higher RH in the marine boundary layer.”

Frank wrote: “And probably cause an increase in marine boundary layer clouds – which can reflect SWR without reducing OLR much and therefore are the most effective clouds at cooling. AOGCMs don’t do a good job with boundary layer clouds, probably because they don’t have the resolution to model to model this interface.”

In the marine boundary layer there is a RH gradient between the surface (usually about 80%) and the top (often 100%). Cloud formation is, I think, strongly dependent on mixing across the top of the boundary layer. So it is not at all clear that higher RH at the surface means more clouds.

Frank appears to be right about the models being at sea with regard to low marine clouds. From AR5, section 7.2.5.3:
“Differences in the response of low clouds to a warming are responsible for most of the spread in model-based estimates of equilibrium climate sensitivity … This discrepancy in responses occurs over most oceans … is usually associated with the representation of shallow cumulus or stratocumulus clouds”

With respect to low cloud feedbacks, they go on to say:
“Starting with some proposed negative feedback mechanisms, it has been argued that in a warmer climate, low clouds will be: (1) horizontally more extensive, because changes in the lapse rate of temperature also modify the lower-tropospheric stability … (2) optically thicker, because adiabatic ascent is accompanied by a larger condensation rate … (3) vertically more extensive, in response to a weakening of the tropical overturning circulation”.
and
“positive feedback mechanisms … warming-induced changes in the absolute humidity lapse rate change the energetics of mixing in ways that demand a reduction in cloud amount or thickness … energetic constraints prevent the surface evaporation from increasing with warming at a rate sufficient to balance expected changes in dry air entrainment, thereby reducing the supply of moisture to form clouds … increased concentrations of GHGs reduce the radiaive cooling that drives stratiform cloud layers and thereby the cloud amount”.

It seems it is complicated. 🙂

22. Angech,

In fact, the basic issue of the nature of the future cloud feedback is not clear. Will it remain negative? If the planet warms, then it is plausible that evaporation will increase, which probably implies that liquid water content will increase but the volume of clouds may not.

Was from the Third Assessment Report (TAR). In what way is this contradictory to the most recent IPCC assessment, AR5:

…a 90% (very likely) range of −0.2 to +2.0 W m–2 °C–1, with a 17% probability of a negative feedback.

It would be helpful if you could provide a link or reference to quotations. I had to google that to find where it was from.

23. Angech,

on your second point, this is hard to parse:

On an if which contradicts the history of cloud formation and negative feedback up to this level of current global temperature [the basic issue of the nature of the future cloud feedback is not clear].
As we both agree the cloud feedback is negative, was zero when it started[no clouds] and has increased up until this current temperature.
In that case there is little need for concern.
To have a problem one needs a positive feedback starting from our current temperature.
Magically, as in we need to believe this to move to step 2, evaporation increases and more water vapour but now at 14.5 C [the current world temp approx] the volume of clouds may not.
Just like that.
Plausible?
To cherry pick a point and say at just this temp all the other processes that went on below this temp suddenly stop?

But I think you are confusing two things:

(1) The total current magnitude of cloud forcing. I will refer to this as “C”. It has units of W/m2.
(2) The rate of change of this forcing with respect to temperature – this is the feedback. I will refer to this as “f”. It has units of W/m2K

It is entirely possible for these to have different signs. Approximately:

C=-18 W/m2 currently (from above posts, I’ve not checked in the literature)
f=+1 W/m2K taking the rough midpoint of the AR5 reference above.

then C=-18+fdT where dT is the difference of a future temperature from the current

Note there is no sudden change at 14.5 degrees, as you claim. No literature makes such a claim – unless you can provide it, which to date you have been unable to.

We do NOT both agree that cloud feedback is negative. It is most likely positive, though the uncertainty is such that it may be negative.

Note that the feedback is defined with respect to temperature (see the units, W/m2K), whereas you sometimes seem refer to it with respect to water vapour.

• “I think you are confusing two things:
(1) The total current magnitude of cloud forcing.
(2) The rate of change of this forcing with respect to temperature
It is entirely possible for these to have different signs.
then C=-18+fdT
Note there is no sudden change at 14.5 degrees, as you claim. No literature makes such a claim – unless you can provide it, which to date you have been unable to.”

I do not think so.
The albedo effect is dependent on the amount of cloud receiving insolation [at different elevations latitudes , longitudes and time of day, independent of the earth temp.
The GHG effect of the clouds is really the GHG effect of the total amount of water in the atmosphere though it is sometimes mistakenly broken up into components and is temperature dependent.
Both are hard to estimate.
Your calculation should be more C= gT-aT where g is the GHG effect on temp of all the water in the atmosphere [ equivalent to the water vapour plus water vapour in clouds as ice etc] and a is the albedo effect dependent on the amount of clouds present and the albedo effect generated.
There are subtle important differences in that the water vapour cannot change much but cloudiness and albedo can increase quite dramatically depending on location and winds, not temperature.
Other effects with rates of water vapour production and ice melt/freeze not considered.

Now there is a sudden change at 14.5 C. The world still warms but faster. The IPCC directs that cloud albedo effects become positive.
-AR5 (2013) “The sign of the net radiative feedback due to all cloud types is… likely positive”
This results in a change of sign,a sudden change at 14.5 C, to the direction of the cloud feedback. It becomes negative.
We have reached peak albedo according to the IPCC.
(2) The rate of change of this forcing with respect to temperature – It is entirely possible for these to have different signs you say means that if one grants negative forcing [positive feedback] to clouds at 14.5 C then there is a sudden change at 14.5 C to the degree of forcing from clouds. This goes negative [downwards] instead of positive [upwards]

• -AR5 (2013) “The sign of the net radiative feedback due to all cloud types is… likely positive”
This results in a change of sign,a sudden change at 14.5 C, to the direction of the cloud feedback. It becomes negative.

This is simply wrong.

The feedback is of the order of 1 W/m2K

This applies whether the temperature change is positive or negative from current.

It is continuous. If the change in temperature is negative, the change in forcing is negative, and vice versa.

There is no sudden change.

I have absolutely no idea why you think there is.

24. VTG
“Let’s go back to your original claims
Firstly This was not in dispute in early IPCC versions but is now considered wrong.” Please provide a citation that demonstrates this.

OK I will try. The paper referencing this is the Zelinka paper mentioned above. Page 676 figure 3

The concept was that there was uncertainty in earlier IPCC findings.
This uncertainty is not in dispute. Plain to see
Shown in the first 4 referenced below

Key statements

FAR (1990) “There is no simple way of appraising the sign of [the cloud amount] feedback” “A further possible negative feedback due to increases in the proportion of water cloud at the expense of ice cloud has been
identified”
SAR (1995) “feedback mechanisms controversial” “the sign of the cloud liquid–water feedback in the real climate system is still unknown”
TAR (2001) “In spite of model improvements “There has been no apparent
narrowing of the uncertainty range associated with cloud feedbacks in current
climate change simulations”“…the sign of [the cloud] cover feedback is still a
matter of uncertainty…”
AR4 (2007)
“…it is not yet possible to assess which of the model estimates of cloud feedback is the most reliable” “The shortwave impact of changes in boundary-layer clouds, and to a lesser extent mid–level clouds, constitutes the largest contributor to inter-model differences in global cloud feedbacks”

But then certainty of a type [ie the uncertainty is now considered wrong]

-AR5 (2013) “The sign of the net radiative feedback due to all cloud types is… likely positive”

• None of this remotely supports your actual claim:

In other words increasing water vapour has a negative feedback effect that is very substantial and at odds with your quote of 2-3 times doubling.
This was not in dispute in early IPCC versions but is now considered wrong.

Yes, science has moved on, clouds are now viewed as more likely having a positive than a negative feedback.

But nowhere have you a quote in the early IPCC reports claiming a “negative feedback effect that is very substantial”, nor one from a later report which contradicts it.

On the contrary, all of these reports claim a large negative forcing from clouds, and a feedback which is uncertain in its sign.

Again, I urge you to understand that forcing and feedback have different units.

Your claim that the IPCC has changed position this was mistaken. You should withdraw it.

25. VTG ” Magically at 14.5 C water vapour which up to that temperature has a known negative feedback turns into a positive feedback, dare I say saving the models? Please provide a citation that water vapour feedback is claimed to change sign at 14.5 degrees”.

“DeWitt Payne angech is not known for the precision of his writing. I would give him a pass on that.”

You win, water vapour feedback does not change sign at 14.5 C. That is not what I was trying to say.
I will try again if you forgive me.
Water in its various forms in the atmosphere, water vapour, clouds, ice crystals etc I have mistakenly referred to at times by the generic term water vapour meaning all water in the atmosphere including clouds.
Water vapour has a known positive feedback which is probably stable since it is mainly the GHG effect.
Water vapour is associated with clouds composed of ice and water droplets.
Even a sky that appears cloud free to the eye has microscopic collections of clouds formed from water vapour in ice or water droplet aggregations which add to the albedo of the atmosphere.
Every hint or trace of cloud increases the albedo hence reducing the global temperature. This is a negative feedback. It is quite strong and outweighs the positive effect of pure water vapour when considered globally.
The fact is that this has occurred up to our current Global Temperature say 14.5C. Yet papers like Zelinka, and commentators like yourself, SOD and Mike M wish to say -AR5 (2013) “The sign of the net radiative feedback due to all cloud types is… likely positive”
Now this statement was not made with certainty previously by the IPCC prior to 2013.
The statement is not physically or scientifically true for all lower temperatures, that is there is a known mechanism of albedo reduction by clouds over 12 hours greater than all GHG effects of water in water vapour and clouds over 24 hours.
Not to mention absolutely swamping CO2 effects.
Yet as I was trying to say, at 2013 magically cloud effects -AR5 (2013) “The sign of the net radiative feedback due to all cloud types is… likely positive” become positive.
I understand that this finding is needed to save AGW as a problem.
I fail to see how all the effects up to 14.5 C in 2013 are albedo> GHG effects and suddenly , post 2013, The sum effect GHG of clouds and water vapour becomes positive that is greater than the increasing albedo effect of more clouds.
I know you can give explanations.
But I want ones that do not stretch credulity.
I hope I have expressed myself a bit better this time. Thank you for correcting my more egregious thoughts and poor explanation.

• Angech,

1) Let us accept that an Earth whose temperature was absolute zero, there would be no clouds, and cloud forcing would be zero.

2) Let us also accept that at some very high temperature, say 5000K, there would be no clouds (no liquid water would be possible), and cloud forcing would be zero.

3) Let us accept that your assertion that current total cloud feedback is negative (I agree with that, as does the IPCC)

4) So across a *very* wide temperature change, there must be a region where feedback is negative, and a region where it is positive, and a temperature where forcing is at a minimum: the forcing vs temperature cure is U shaped. Let’s call the temperature of the minimum Tmin, with cloud forcing equal to -Cmin.

5) For all temperatures T<Tmin, feedback must be negative (cloud forcing decreases with increasing temperature, from zero at T=0K)

6) For all temperatures T>Tmin, feedback must be positive (cloud forcing increases with temperature from -Cmin at T=Tmin up to zero at 5000K)

7) In this framing of the problem, the IPCC are saying is that it is likely (though not certain) that current surface temp is >Tmin.

I hope that helps.

3) Let us accept that your assertion that current total cloud forcing is negative (I agree with that, as does the IPCC)

• Totally in agreement with you here.

• Totally in agreement with you here.

Hooray! I’m very tempted to leave it at that… however… you do realise this is in contradiction to your earlier posts??

• vtg,

At high temperature, much, much less than 5000K, all carbon will be in the atmosphere as CO2 and water will be gone too. But that is extremely unlikely with the sun at it’s current stage. I believe it was Ramanathan who stated that as long as there is a hole in the atmospheric spectrum allowing some surface radiation to escape directly to space, a runaway greenhouse is not possible. The surface temperature would have to go up to something like 50°C to close the hole.

I’m still skeptical that there is enough data (model runs are not data, IMO) to make the statement with any confidence that the sign of cloud feedback is likely positive.

• Payne,

sure, for 5,000 read “any temperature high enough to prevent liquid water”.

To be clear, I am not remotely suggesting such a temperature is attainable, let alone likely, merely that in principle if it were attained then cloud feedback would be zero, therefore there must somewhere be a minimum. Just a thought experiment.

I don’t have the expertise to argue what the right number for cloud feedback is. My intuition says that overall climate sensitivity cannot be low, or the glacial cycle would be damped out of existence. But that’s a crappy, qualitiative argument. Happily, though, it seems to be backed up by experts.

https://www.nature.com/articles/nature11574

I’m not going to pretend that in reality I have a better insight than the IPCC do.

As a writer of heat transfer textbooks, you surely have a much greater claim to such insights than I do 😉

• vtg,

That’s a different Payne. I haven’t written any books.

• vtg,

My intuition says that overall climate sensitivity cannot be low, or the glacial cycle would be damped out of existence. But that’s a crappy, qualitiative argument.

Conversely, overall climate sensitivity can’t be too high or the system would have run away to one extreme or the other. Also a crappy qualitative argument. You don’t have to have much sensitivity if ocean circulation changes, such as the closing of the Isthmus of Panama, creates a small polar heat leak resulting in an imbalance that adds up over hundreds of thousands of years.

• “Every hint or trace of cloud increases the albedo hence reducing the global temperature. This is a negative feedback.”

Angech, did you ever read my comment referring to Lindzen’s “iris hypothesis”? It proposes a negative cloud feedback as earth is warming…by reducing cirrus cloud cover. That is, following your logic, less cloud cover = less albedo = increase in global temperature. However, Lindzen’s hypothesis stated the exact opposite in the last part: it would result in a reduction in global temperature.

Now, the consensus view at the moment is that warming will increase cirrus cloud cover and thereby increase global temperature – again exactly opposite to your view.

https://en.wikipedia.org/wiki/Cloud_albedo

It simply isn’t as simple as you think it is.

• Nice try, two different concepts.
How to put it.
If you have something white that reflects light and you increase that it will reflect more light, energy, SW whatever.
Always a negative feedback
Clouds at different heights do alter the radiative balance of whatever heat , light energy etc got through positive feedback on a smaller amount of energy.
Ones further out strangely cop a lot more IR on the way in and out so have a higher positive GHG feedback, I think?
Negative cloud feedback and negative feedback on clouds are two very different animals.
You will have to ask yourself why Lindzen comes up with a seemingly contradictory hypothesis, he would have good scientific reasons but nothing in it contradicts the physics of the albedo effect of increasing clouds.
People claiming magical turning points often have to resort to less cloud formation at turning points, as you will already have noted in the comments.
Show us the science that makes clouds at all lower temperatures as the temp increases and the reasons why it should suddenly stop at the very temperature we are at now.
Interested.

26. That’s a different Payne. I haven’t written any books.

That’s very honest of you! You could have got away with that.

27. ” I doubt that anyone who has troubled themselves to read even one paper on climate modeling basics could reach the conclusion so firmly believed in fantasy climate blogs and repeated above. If you never need to provide evidence for your claims.

Well there is Cess, “Cloud feedback in atmospheric general circulation
models: An update” FAR 1990
changing parameters at will.

‘Since this change alone made the models much darker than
satellite observations indicate, they were then brought back into
general agreement with observations by altering the critical
relative humidities for cloud formation and other modification”

28. verytallguy Totally in agreement with you here.
“Hooray! I’m very tempted to leave it at that… however… you do realise this is in contradiction to your earlier posts??” Not at all. You say,

4) So across a *very* wide temperature change, there must be a region where feedback is negative, and a region where it is positive, and a temperature where forcing is at a minimum: the forcing vs temperature cure is U shaped. Let’s call the temperature of the minimum Tmin, with cloud forcing equal to -Cmin.
5) For all temperatures TTmin, feedback must be positive (cloud forcing increases with temperature from -Cmin at T=Tmin up to zero at 5000K)

I said “”we both agree the cloud feedback is negative, was zero when it started [no clouds] and has increased up until this current temperature.
To have a problem one needs a positive feedback starting from our current temperature”
Magically, as in we need to believe this to move to step 2, evaporation increases and more water vapour but now at 14.5 C [the current world temp approx] the volume of clouds may not.”

So pick a magical point at T 14.5C
“Let’s call the temperature of the minimum Tmin,” OK.
Well we do have boiling points, freezing points and now the magical adiabetic relative humidity Cloud puncturing point MARHCP point.

You do realise that the IPCC and you are calling 14.5C the turning point.
After all all 19 models used by the IPCC predicted positive cloud feedback at this point.

• angech wrote: “You do realise that the IPCC and you are calling 14.5C the turning point.”

Nonsense. As verytallguy said in his extremely clear explanation, the (uncertain) implication of the models is that the present temperature is above the turning point (actually, the minimum), not at it. But the point is that it is a minimum which could quite possibly be extremely broad.

• “the point is that it is a minimum which could quite possibly be extremely broad.” might as well say
“the point is that it is a minimum which could quite possibly be extremely short”

“As verytallguy said, the implication of the models is that the present temperature is above the minimum”.

No, he said nothing about the implication of the models.
He said
“3) Let us accept that your assertion that current total cloud feedback is negative (I agree with that, as does the IPCC)”
-But all 19 models used by the IPCC predicted positive cloud feedback at this point 1990. see Cass. See

4)there must be a region where feedback is negative, and a region where it is positive, and a temperature where forcing is at a minimum: the forcing vs temperature cure is U shaped.
5) For all temperatures TTmin, feedback must be positive (cloud forcing increases with temperature from -Cmin at T=Tmin up to zero at 5000K)

7) In this framing of the problem, the IPCC are saying is that it is likely (though not certain) that current surface temp is >Tmin.
Therefore it is also likely (though not certain) that current surface temp is <Tmin.

Given that all 19 models used by the IPCC predicted positive cloud feedback at this point 1990. and again in 2013

Given that all 19 models used by the IPCC predicted positive cloud feedback at this point 1990. and again in 2013 the IPCC models and you are calling 14.5C the turning point.”

Thanks for trying to flesh out this problem, Mike M.
The question of where the minimum lies is very important even if we differ.
-VTG said
5) For all temperatures TTmin, feedback must be positive (cloud forcing increases with temperature from -Cmin at T=Tmin up to zero at 5000K)
and
7) In this framing of the problem, the IPCC are saying is that it is likely (though not certain) that current surface temp is >Tmin.

So though he said “3) Let us accept that your assertion that current total cloud feedback is negative (I agree with that, as does the IPCC)”
in 7] he said the IPCC was saying cloud feedback was positive as >T min.
In other words we had passed Tmin currently, his words.

• Angech,

If 14.5 degrees was the turning point, cloud sensitivity would have to be *zero*. That is not claimed.

Also, cloud sensitivity would have to be claimed as *changing* with temperature. That is not claimed.

You do realise that the IPCC and you are calling 14.5C the turning point.
After all all 19 models used by the IPCC predicted positive cloud feedback at this point.

are in direct contradiction of each other. They are oxymoronic.

• verytallguy
“If 14.5 degrees was the turning point, cloud sensitivity would have to be *zero*. That is not claimed.”
“4)there must be a region where feedback is negative, and a region where it is positive, and a temperature where forcing is at a minimum: the forcing vs temperature cure is U shaped.”

“Also, cloud sensitivity would have to be claimed as *changing* with temperature. That is not claimed.”
“5) For all temperatures TTmin, feedback must be positive (cloud forcing increases with temperature from -Cmin at T=Tmin up to zero at 5000K)”

I apologize if I am mixing up cloud sensitivity and cloud forcing like I mixed up water vapour and clouds made out of water vapour[as ice etc].
I believe there are GHG radiative and albedo effects combined in working out the forcing or feedback.
Nonetheless the idea of what we are talking about seems to be on the same plane that you postulate and I accept that if clouds are to have a positive feedback there is a turning point at Tmin where the clouds reverse their known effect, up till now, on global temp and if the models are to be believed that must be now?

Your last two sentences are oxymoronic and in direct contradiction of each other?

– You do realise that the IPCC and you are calling 14.5C the turning point.
“3) Let us accept that your assertion that current total cloud feedback is negative (I agree with that, as does the IPCC)”
Yet all 19 models used by the IPCC predicted positive cloud feedback at this point. [Fact]

Hence complement each other perfectly if we are indeed at your Tmin at 14.5 C.

29. The statement in AR5 on cloud feedback that “No robust mechanisms contribute negative feedback” is not a particularly convincing argument. Back when I was an undergraduate, continental drift was not accepted, particularly in academia in the US. One of the arguments against continental drift was basically the same. Continents couldn’t move because there was no mechanism to cause them to move. Or at least there wasn’t until there was.

30. Unrelated to this post, I have a question.

What is the resolution of the reliability of the past temperature data we have for time periods very long ago? Are our temperature records accurate to within 100 years, or 50, or 10? I know that ice cores and tree rings are used, but my expectation is that they’d be more reliable for recent years and that they might not work well from year to year even if they work well on broader timescales.

I don’t expect this would matter very much, but it might matter a little.

• 27chaos
“What is the resolution of the reliability of the past temperature data we have for time periods very long ago”
An interesting point in this question is that as you go further back in time the possibility of large seasonal swings in temperature may occur but would not be detected in the compacting together of data much further in the past.
Reliability is measurement and correlation dependent. If you can tie two or 3 different proxies together in the same area the records become more reliable.
The further back in time you go the less reliable they are.
Tree rings suffer from other factors than temperature e.g rainfall, incident light and substrate composition that they grow on, bit like my garden really.

31. Angech,

I’m now going to give up here.

Your latest posts continue to be absolutely confused and illogical on the science; it’s clear there is no purpose in my trying to help you understand it.

You posit that the IPCC summary of the science is magical thinking.

Ask yourself how likely this is, what alternative explanations there might be for your disagreement with them, and how likely those explanations might be.

32. VTG
use of the m word was uncalled for.
I do think we are in agreement re Tmin as a concept.
I am unclear where you think it might be.

33. SOD
“If you read through you see that the emission of radiation at a given wavelength is only dependent on the temperature of the local gas and the emissivity at that wavelength.
Usually we get visitors who haven’t read textbooks showing up and claiming the opposite – that when a CO2 or water vapor molecule absorbs a photon it re-emits it. And that the energy absorbed is not absorbed into the average temperature of the local gas by collision. Then we have to explain the average time for collision in gas at a given pressure is 1000x shorter than the time to re-emission of a photon for this gas molecule in an excited state.”

If you have time could you kindly clear up a little confusion on my part.
The emission spectrum of earth from space shows bands of missing IR from CO2 and water. This implies that the CO2 in particular must re emit at different temps to that it absorbs at. Would this not lead to a spike in emission at the emitting temperatures given that they have removed the energy at the absorption level?
This also leads on to a question of what the spectrum appears to be on earth looking up or down. In particular looking up should be the same as looking down, or not?

• angech,

The missing bands don’t show that at all.

If a gas absorbs at a wavelength it can also emit at that wavelength. If a gas doesn’t absorb at a wavelength it cannot emit at that wavelength.

The strength of absorption (called the absorptivity) equals the strength of emission (called the emissivity).

• angech,

The short answer is that emission to space from ghg’s comes from higher in the atmosphere where it’s colder, so emission intensity is less than at the surface.

Another good reference is Grant W. Petty, A First Course in Atmospheric Radiation. It’s available directly from the publisher, Sundog Publishing, for US \$36.

If you’re too cheap for that, try :

For looking at emission spectra with different conditions and observation altitudes and directions (up or down) there’s this:

http://forecast.uchicago.edu/Projects/modtran.html

34. Thanks, will get back when I am more informed, obviously not understanding the reason for the CO2 etc signatures yet

35. Article up at WUWT on radiative physics, seems should clearly be wrong. Any comments I should take back to that forum?
In particular can a CO2 molecule absorb and emit radiation only because it is at a certain energy level itself or does it just do in case of absorption because it encounters that wavelength and what level is it before and after absorption /emission?

• Ron,

In a word, bogus. The author clearly has no understanding of quantum mechanics. This is demonstrated by his comment that electrons orbit the nucleus at fixed altitudes. At the atomic scale, there is no such thing as an orbit or a fixed altitude. That was from a skim on the first page. I have absolutely no interest in a detailed reading and analysis. I know for a fact that whatever his conclusions, if they do not agree with quantum electrodynamics, they’re wrong.

36. rH is decreasing in the Troposphere over decades, shown in weather or climate models? https://www.friendsofscience.org/index.php?id=710

37. on May 12, 2020 at 9:03 pm | Reply nobodysknowledge

Scientists and climate models seem to underestimate some water vapor negative feedbacks.
A new study from Seth D Seidel and Da Yang show this: The lightness of water vapor helps to stabilize tropical climate. May 2020.